Skip to content

Conversation

@rniczh
Copy link
Contributor

@rniczh rniczh commented Oct 20, 2025

Context:

Description of the Change:

Benefits:

Possible Drawbacks:

Related GitHub Issues:

@rniczh rniczh changed the base branch from main to rniczh/bump-jax-to-0.7.0.dev20250703+cd1b9520b October 20, 2025 19:14
@rniczh rniczh changed the base branch from rniczh/bump-jax-to-0.7.0.dev20250703+cd1b9520b to rniczh/update-llvm-version-20251009 October 27, 2025 03:49
@JerryChen97
Copy link
Contributor

Fantastic, now even Codecov passes! @rniczh

**Context:**
Basically, for Catalyst, the new `bump-jax-patching` branch means
there's no patching happenning by default anymore.
In this way, we can better separate the impact of patching, instead of
causing inheritance of patching from upstream to downstream.
We could also better observe what patches are actually needed for other
packages depending on Pennylane.

**Description of the Change:**

**Benefits:**

**Possible Drawbacks:**

**Related GitHub Issues:**
@JerryChen97
Copy link
Contributor

Now Catalyst's jax bumping branch should be independent of whether the PL Jax patches are global or local.
Still need https://github.com/PennyLaneAI/pennylane/pull/8525/files due to hashability issues as well as API changes regarding makevar, new_jaxpr_eqn, and getvar

Copy link
Contributor

@JerryChen97 JerryChen97 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hold on temporarily until the PennyLane PR is also ready-to-merge

@JerryChen97
Copy link
Contributor

@rniczh Merge?

@rniczh rniczh merged commit 0b95842 into main Nov 28, 2025
39 checks passed
@rniczh rniczh deleted the rniczh/bump-jax-to-0.7.0 branch November 28, 2025 20:41
maliasadi added a commit that referenced this pull request Dec 1, 2025
Revert PR #2215 changes that accidentally merged in PR #2131.

Note that these changes don't break the current functionalities of Catalyst,
but they introduce edge cases with lowering generic PL operators in MLIR
that need to be properly addressed before the merge.
(re: the future Operator class)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

6 participants